I am head of the Theoretical Crystallography Group. My research is concerned with the determination of molecular structure using the diffraction of X-rays, electrons and neutrons by crystalline materials. In a diffraction experiment, crystals are irradiated with such radiation, and interference effects give rise to a characteristic diffraction pattern which is a unique to that crystal. This pattern has two features - the intensities of the diffracted beams, and their associated phases. However, whereas the intensities are measurable the phases are not, and this is the phase problem of crystallography - one of the fundamental problems of structural science. I use direct methods, usually based on maximum entropy, to recover these phases and apply these techniques to problems in:

In all these cases not only is there a phase problem, there is also a problem with the quality of the data – it is often incomplete and prone to error. This makes the maximum entropy method especially useful since it is a very robust technique.

I also work on problems associated with the matching of powder diffraction patterns and the quantitative analysis of mixtures i.e. determining the components of a mixture by examining and analysing their powder pattern. This work is based around the development of the SNAP-1D computer software. Further development of SNAP-1D lead to the development of PolySNAP, allowing high-throughput analysis of up to 1500 powder diffraction patterns in a single run combined with a flexible graphical interface to summarise results. PolySNAP is marketed and distributed by Bruker-AXS.

Currently in development is the clustering program dSNAP. This program applies clustering methods to data mined from the Cambridge Structural Database and displays the results in an easily interpreted way. This allows the user to interpreted the underlying Chemistry and takes the leg work out of structural comparisons.